Motor Controller/VFD/Soft-Start Sizing

[mayanees]

I did design for a 50 HP Soft-Start on a submersible pump, sizing the controller and cable for 50HP, based on the motor amperages in 430.250 - Full-Load Current, Three Phase Alternating -Current Motors.

The motor draw has been measured to be 77 amps @ 460V, equating to 60 HP per NEC 430.250 - resulting in death of the soft-start.

The civil engineering firm that gave me the 50HP number says the town may be asking my company for reimbursement to upgrade the controller to 60 HP. I tell him that my design is adequate for 50 HP, as directed. And, if he was confident in his calculation of 50 HP, look for some other problem like pipe restriction on the outlet that may be causing high power requirement. Otherwise, he underestimated the pump motor requirement.

And the only thing that has me doubting the basis of the design is "Service-factor" of a motor. If the motor has a 1.2 SF, that would cause the controller to see 60 HP of load. I think this one has a 1.15, (I never saw the motor submittal to verify nameplate data), and I think the 50 HP Soft-Start may have handled that. But this measured load is at 118.5% of fla.

I always size everything for "minimum-essential", that being the smallest size that adequately meets the criteria. To me that's good-engineering not to over-design at the expense of a client. The tables are there to be "sized to".
In this case if I were made aware that the load was going to be 60 HP, I would have sized to that, and there wouldn't have been a problem.

Thanks for any commentary.

JM

 

[kingpb]

I do not agree with your thought that, "The tables are there to be sized to", . Read NEC 90.1, the Code is for safety, only.

As far as the current draw; what is the motor nameplate Hp? If it is 50Hp, but drawing 77A, that is one situation that needs to be dealt with, but if the nameplate is 60Hp, and you were directed to design for 50Hp, then ask the civil engineer for a copy of the documentation whereby the motor size was change.

The soft starters I am familiar with, are based on current capability not Hp, and therefore any service factor that might cause the current to be greater then the FLA rating/NEC tables should have been considered for selection of the soft starter. In this case, it may have been that the starter size was borderline, and due to the situation, the motor is operating in the S.F. for long periods of time, thus exceeding the current rating of the starter.

This may have to fall into the category of "Lessons Learned"

 

[mayanees]

90.1

90.1

That's a point well taken. The table's a minimum, but it's surely the starting point for the design.
As far as the motor nameplate, it's 50 HP. I believe the motor is operating well into and over its nameplate rating, and that's the problem.
One of the civil guys told me that they calculate the expected HP, but don't really know what it's going to be until they try it, implying that they should have spec'd a 60 HP motor and controller.
I advised them to upgrade to a 60 HP soft-start, and run the motor until it fails, at which time they should replace it with a 60 HP motor. The cable is sized to accommodate 60 HP, since I upsized for voltage drop.
JM

 

[Jraef]

I hereby set myself up as an impartial judge, jury and executioner.:grin:

Your culpability in this:
Lazy engineering. You should have known to CHECK, CHECK and CHECK the facts and data. What was the motor nameplate FLA? That is the ONLY number that you should have responded to. Forget HP, forget tables, forget everything else. You should have waited until you personally saw the pump submittal, together with an expected current draw at the flow they were selling to, before finalizing the design of the electrical equipment. If you were required to engineer it without that information, you should have made very clear disclaimers UP FRONT to all involved parties that without that information, you can only ESTIMATE the proper equipment sizing.

The pump supplier's culpability:
Shoddy brinkmanship engineering. It is an old trick of the pump supplier industry to overreach the capacity of a motor by using the Service Factor rather than go one size up. A 50HP motor should have had no more than 65A, +- 5% depending on design, speed etc. So in that regard, I think you are correct in that they are running into the SF. They can do what they like, but at the very least they should have been forthcoming with that information. NO motor control equipment is going to have an 18% "fudge factor" in it's sizing. More importantly though, what they are ignoring is that by doing so they are significantly shortening the life of the motor at the end user's expense, because it will essentially last out the warranty on the pump but fail when the user has to bear the entire cost. NEMA MG-1 design specifications will back that up.

Judgment: For the defendant (you).
Although you are guilty of being lazy, that does not imply criminal intent and financial responsibility for the sake of profit as the actions of the pump supplier do.
The responsibility for pump performance lies squarely in their camp. If the pump could not perform at 50HP and they needed to use the SF, it was their responsibility to let you know that, but what really needed to happen was for them to insist that the original design was wrong and it needed a 60HP pump. Running continuously in to the SF is not an accepted engineering practice.

Sentence:
50 lashes with a wet noodle for you,
Cost to upgrade to a 60HP controller to the pump supplier.

Court is adjourned, have a nice day.

 

[weressl]

I hereby set myself up as an impartial judge, jury and executioner.:grin:

Your culpability in this:
Lazy engineering. You should have known to CHECK, CHECK and CHECK the facts and data. What was the motor nameplate FLA? That is the ONLY number that you should have responded to. Forget HP, forget tables, forget everything else. You should have waited until you personally saw the pump submittal, together with an expected current draw at the flow they were selling to, before finalizing the design of the electrical equipment. If you were required to engineer it without that information, you should have made very clear disclaimers UP FRONT to all involved parties that without that information, you can only ESTIMATE the proper equipment sizing.

The pump supplier's culpability:
Shoddy brinkmanship engineering. It is an old trick of the pump supplier industry to overreach the capacity of a motor by using the Service Factor rather than go one size up. A 50HP motor should have had no more than 65A, +- 5% depending on design, speed etc. So in that regard, I think you are correct in that they are running into the SF. They can do what they like, but at the very least they should have been forthcoming with that information. NO motor control equipment is going to have an 18% "fudge factor" in it's sizing. More importantly though, what they are ignoring is that by doing so they are significantly shortening the life of the motor at the end user's expense, because it will essentially last out the warranty on the pump but fail when the user has to bear the entire cost. NEMA MG-1 design specifications will back that up.

Judgment: For the defendant (you).
Although you are guilty of being lazy, that does not imply criminal intent and financial responsibility for the sake of profit as the actions of the pump supplier do.
The responsibility for pump performance lies squarely in their camp. If the pump could not perform at 50HP and they needed to use the SF, it was their responsibility to let you know that, but what really needed to happen was for them to insist that the original design was wrong and it needed a 60HP pump. Running continuously in to the SF is not an accepted engineering practice.

Sentence:
50 lashes with a wet noodle for you,
Cost to upgrade to a 60HP controller to the pump supplier.

Court is adjourned, have a nice day.

I object, your honor. (I don't know what to but I object on principle.)

Let me clear up something. It is never my fault, ever, or anyone whom I may be defending. Get that into the grey matter under your powedered wigs. Ooops, wronmg court. Nevermind...

The submersible pumps are a branch of different species. NEMA MG-1 is very likely be not even applicable, nor are the NEC tables.

Their construction is very different from ordinary squirrel cage motors as th stator and rotor are much "skinnier". As the result the inrush current will be nothing like ordinary motors. Accordingly; the protection, voltage drop calculation and wire sizing have to be custom made not just from the nameplate, but other relevant data on inrush and motor load curve. The full load current will also be different since the motor is expected to operate at a much lower ambient than the customary 40Centigrade, so in essence you can get 60HP or even more out of a motor that would be rated 50HP under ordinary service. (This issue routinely comes up in our industry in reverse, with canned motors that have the winding exposed to the process fluid of 100Centigrades or even higher.)

So live and learn...

As a side comment. ASD's being less and less expensive - espoecially wehn compared to a "soft starter" - and more oriented toward demand, this application might have been better served with pressure signal controlled ASD instead of a SSRV.

 

[mayanees]

my response

my response

I didn't much appreciate the charge of being lazy, because I'm anything but. And I hung myself out on this forum to get assistance with a real problem, not to be toyed with.

But I do appreciate the latter response which stated:

The submersible pumps are a branch of different species. NEMA MG-1 is very likely be not even applicable, nor are the NEC tables.

... because I think that's the root of the problem, and I want to know how to handle it technically. I don't feel comfortable automatically adding 20% capacity to the starting means for a submersible pump - unless that's the way it's done. I've done designs for about 15-20 of these over the past 10 years, and this has never been a problem.

In my EE mind, power is power. It takes a certain HP to get the water out of the well. In this case, the Civil engineer estimates that it takes 50 HP, and is surprised to see that it's taking 60 HP. But the fact of the matter is that in order to perform the work required to get the water up it's taxing the motor at 60 HP. If the work to be performed was less than 50HP, i.e. if the well were more shallow, the motor would not pull as much power.

Having experience with 15-20 other wells where this wasn't a problem, I'm reluctant to overdesign every submersible because they can run deeper into their service factor.
My opinion in this situation is that it's taking more HP than the calculations showed - and not that a 50 HP submersible pulls 60 HP.

Any objective responses are appreciated.

JM


...edited for spelling.. further evidence that I'm not lazy

 

[mdshunk]

Here's my summary and commentary...

You gave them a 50 horsepower total system, as they requested.
It doesn't work, because they need 60 horsepower.
You're not a mechanical engineer or fluid dynamics guy, so how were you supposed to know that 50hp motor would be digging into its service factor so much?

If it was me in your shoes, I'd tell them about how many engineering hours you'd have involved in designing a 60 horsepower system, and about how much that engineering would cost them.

 

[micromind]

I'm not an engineer, just a journeyman, so I'll look at it from that angle.

First, what caused the soft-start to fail? Does it have a by-pass contactor in it? Does it actually work? If not, it will produce considerable heat. If it's in a non-ventelated enclosure, it'll overheat quickly. Even so, the heat-sink overtemp's should have caught it, and shut it down before it was destroyed.

Look carefully at the control wiring, most soft-starts have their own internal power supply, and applying outside voltage to the control inputs will cook the board. Not always right away.

I've never seen a submersible pump (cold water) that didn't load the motor well into the service factor, but even that shouldn't have caused the soft-start to burn up. I'd look real hard at why it failed, there may well be nothing wrong with the design.

If the current was significantly imbalanced, try 'rolling' the phases. This makes absolutely no sense at all, but it usually works. Say you're using brown-orange-yellow for A-B-C. Move brown to B, orange to C, and yellow to A. Then check current. Then brown to C, orange to A, and yellow to B. Check current again. The rotation is the same, but one of these combinations will usually result in better current balance. This works best on submersible pump motors, and A/C compressors.

 

[RayS]

And, if he was confident in his calculation of 50 HP, look for some other problem like pipe restriction on the outlet that may be causing high power requirement.
JM

actually, if there was a restriction, it may actually lessen the draw- more head, lower flow. You didn't mention when the 77A? was measured- is this routine, or just at startup?

Sub. pump has a couple of challenges that should have been addressed (by others, like the sizing)- backspin after turnoff (big trouble starting into that) and unloaded operation/higher load if the column has drained. Both challenges can be solved if there is a functional check valve.

 

[mayanees]

... late night response

... late night response

... comments on my dilemma.. going from memory of my read of the latest posts..
mdshunk; you're my new hero. My first girlfriend was a MacDonalds fry technician, so maybe that's where we connect. And you hit the nail on the head with my knee-jerk response to the civil firm: we have responded to all of your inquiries into the overload situation..... and will need to get compensated for the redesign of a 60 HP controller..

controls guy: micromind I think... A reputable controls guy did the t/s and repair of the 50 HP soft-start, actually a professional cohort/accomplice on previous jobs. He kinda buried me and maintained that his experience was that all submersibles pull more power than nameplate. He and I worked through the changes, and I concurred that based on the power measurements, the controller needed to be upgraded to 60 HP components. I honestly don't know the details of the failure, other than "tripping out on Soft-Start overloads.

And to the Austin Powers guy: I flinched at the "lazy engineer" remark, and didn't give much thought to the content of your suggestion because of your delivery. But later that night as I was driving to pick up my 15 year-old from her lifegaurding job, I got to thinking of the basis of your argument. If the civil engineer who spec'd that motor looked at a performance curve for a pump that delivered his required flow and pressure and was listed at 50 HP, then shame on the pump supplier, whose 50 HP pump needs 60 HP to do its work.

Thanks for the interest this has drawn, and keep the comments coming! At this point I'm certainly better equipped to handle this in my next design, but I would still benefit from more opinions.

JM

 

[coulter]

There's something left we don't know.

1. Soft starts are programmable for starting current. A motor with an inherently high inrush won't matter. The current limits at the programmed value.

2. Some soft starts are also programmable for running current - overload. If not, then there should have been seperate overloads installed. Either way the overloads should have been set for the motor FLA and service factor.

So, how come it burned up the soft start? It was pulling too much current, why didn't it shut down?

What you are describing is no different than a motor with dragging bearings. I certainly would not expect a motor with dragging bearings to hose the soft start.

So, what don't we know yet?

carl

 

[frenchelectrican]

I dont know if any one noted this or not if my memory serve me right that if the well head[ pump ] is mounted too shallow for this hp catory or water flowage is more than what the factory spec is ??

and i dont know [ correct me on this one ] did anyone try to " throttle " the water discharge ?? if you can throttle down some the current will drop because of centifuge impeller design.


the other thing is the soft start system probly not programed right and most soft starter useally i know they only can handle 350% of FLA current for few seconds anymore than that it can really cook that unit.

and the other thing i dont know if this will be related but one person did mention the SF [ service factor ] most submerable well motors have sf high as 1.85% [ for smaller units but larger one kinda like 1.35% the most ]

hope this will help some

Merci , Marc

 

[kingpb]

As an earlier post suggested, however submersible pumps are covered by NEMA MG-1. It could possibly be the torque characteristic design that is causing a problem.

But this post brings up a issue, as an engineer, I have always had with the motor FLA tables. In NEC Article 90 it is made clear that NEC is not intended for design, or anything for that matter except safety. Yet, the motors sections, specifically, says you have to use the Tables for FLA, even if nameplate is known. So, am I supposed to base my design, current in this case, on a value out of a NEC Table that could be potentially lower then what is called needed?

Isn't the NEC dictating design in this case, which clearly is a contradiction to NEC Article 90. I cannot speak for others, but as an engineer, responsible for a design, I will always use nameplate data, when available, and use the NEC Tables as a default if no other/better data is known. Do the Tables work in most cases, probably, but the should not be used blindly, without doing due diligence.

I think too often the NEC is used as a design tool, it is NOT, it is for providing a safe installation.

 

[ptonsparky]

My local village had a similar problem. New submersible was continually tripping overloads. 40 hp was trying to do the work of a 60. Total head was greater than what was engineered for. No one made the corrections needed after the initial installation. Pump installers know how to fix this if they are inclined. Bottom line on this one, the engineering firm failed in design and follow up. I wish my bill was going to them instead of the Village.

Pump installers use the service factor because total head is not static. It changes with the water level in both the well itself and the water tower. Unfortuantely they rely on service factor more than they should.

Think of it this way. It is harder and more work to actually lift 450 gallons of water 500 feet, then it is to spin an impeller that is not moving water.

 

[mull982]

As a fairly new engineer this is an interesting topic to read about. "Service Factor" is a topic that I have seen from time to time in dealing with motors and was hoping that someone could clearly explain what it was exactly. My vauge understanding of it is that sometimes a motor has to produce more output than its rated h.p. and it can safely produce more output up to its rated service factor but only for short periods of time.

From reading the previous posts, it appears that when a motor operates into its service factor it draws more current than FLA. With this being the case, when designing feeders for typical induction motors should this service factor be taken into place when calculating the total amp draw of the motor or sizing the ampacity of your feeders. In other words, should you use the FLA of the motor plus the additional service factor amps when designing feeders for a motor?

 

[kingpb]

The following is from NEMA MG-1:

View attachment 857

View attachment 858

Commonly many are mis-informed and think that motors cannot operate in the service factor, continuously. As you can see, this is not the case, although there may be negative aspects to doing so.

 

[winnie]

Service Factor: horsepower is a unit of power, exactly equal to some number of watts. You can rate a motor in terms of its _continuous_ output power, the number of mechanical watts that it can push down the shaft on a continuous basis without overheating.

However the 'horsepower' rating of a motor is more of a nominal term, a composite number that describes some combination of starting torque, break-away torque, full load torque, and various other parameters.

As I understand it, if the continuous _steady state_ output capability of the motor exceeds the composite rated horsepower, then 'service factor' may be used to describe this. For example, a motor with the starting starting torque capability that gets a 50hp rating, but which can be safely run on a continuous basis with 60hp output would get a SF of 1.2

For a particular example: if an induction motor is designed with a very low resistance rotor, it will have higher operating efficiency but lower starting torque. The higher operating efficiency means lower heating and thus greater continuous output capability, but the lower starting torque would mean a lower nominal HP rating. Such a motor would reasonably have a very high SF.

This is just my rough understanding, open to correction or expansion!

-Jon

 

[Jraef]

The following is from NEMA MG-1:

View attachment 857

View attachment 858

Commonly many are mis-informed and think that motors cannot operate in the service factor, continuously. As you can see, this is not the case, although there may be negative aspects to doing so.

Thanks for that attachment.
The last line in that section of NEMA MG-1 was added only in the recent past. Before that it was a lot more vague, leading people to believe they could do it without consequence. That is why NEMA added that line, there IS a consequence, and as I said, the cost is borne by the end user because usually the motor will last out the warranty.

Jon,
Motor design HP is not as complex as you make it out to be, at least not at the motor end; it is just 746 watts, thats all. Starting torque is not part of that equation, although running (full load) torque and base motor speed is, just as the kW rating in IEC motors is.

What you are describing is more about how a motor is selected for a load. When an ME wants a motor, he looks at his load torque and speed requirements based upon the mass he needs to move, then calculates a BHP for that load. The motor selection is then based upon that BHP as a minimum, but may need to be higher if the starting inertia is higher than what the motor torque can overcome. But that just may mean that if he needed 12BHP for the load and a 15HP motor does not provide enough torque to overcome the starting inertia, he may need a 20HP motor. The motor designs for 15HP or 20HP are relatively standardized.

"Service Factor" is nothing more than a fudge factor for short term loading beyond the original design specs.

 

[RayS]

As a fairly new engineer this is an interesting topic to read about. "Service Factor" is a topic that I have seen from time to time in dealing with motors and was hoping that someone could clearly explain what it was exactly. My vauge understanding of it is that sometimes a motor has to produce more output than its rated h.p. and it can safely produce more output up to its rated service factor but only for short periods of time.

yep, pretty close. The motor will run indefinately at the service factor, but with a much shorter life. If memory serves, adding 10deg. C to the operating temp cuts life in half.

The pump is a good example of the need for a service factor, to cover occasional operation into low or no head conditions until the pressure gets up into operating range, where the motor power output will decrease to normal.

Sounds like this particular motor is just plain undersized.

 

[Jraef]

mayanees,
By the way, I'm sorry if my form of tongue-in-cheek humor offended you, I was just trying to be supportive in a ribbing sort of way. Pardon me.

It looks as though you got the gist of my meaning though: you were not at fault here. You did what you were asked, they just asked you do do something that turned out to be inadequate; unbeknownst to you.

As to the soft starter burning up, I still think that even with a bypass contactor, an 18% continuous overload is more than most soft starter manufacturers build-in to their sizing scheme. A quick scan of 5 of the major manufacturers puts 50HP soft starters as being sized for between 64 and 68A (Allen Bradley doesn't offer a 50HP, they jump from 40 to 60). So 77A continuous was still an overload of at least 13%. If you had an oversized external bypass contactor (most of them are now integral) it may have worked, but again, you would have had to know all these details IN ADVANCE, which is what they failed to do.